Chinese AI stocks fall as Nvidia CEO resignation sparks export fears

Market reaction to leadership shift

The resignation of Nvidia CEO Jensen Huang has sent significant shockwaves through the global AI industry. Following revelations of personal scandals involving a suspected Chinese spy, the chipmaker entered a period of high instability. This leadership vacuum has caused immediate volatility in the Chinese stock market.

Investors reacted by selling off shares in prominent Chinese AI model companies. Stocks for Z.ai and MiniMax Group fell sharply as the market weighed the implications of Huang's departure. The uncertainty surrounding Nvidia's future direction has created a ripple effect across the broader tech sector.

While AI model stocks struggled, the market showed a divided reaction. Data center suppliers like Zhongji Innolight and Eoptolink Technology slumped because their businesses rely heavily on US markets. Conversely, some domestic chip stocks gained as investors began to bet on the necessity of Chinese tech self-reliance.

The threat of tighter export controls

The primary concern for analysts is that Huang's resignation will remove a key opponent of strict chip export controls. As one of the most influential voices in the industry, his absence may embolden lawmakers to take a harder line. This shift could lead to a significant reduction in the flow of advanced hardware to Chinese firms.

US House Speaker Mike Johnson noted on Thursday that there is a growing consensus in Congress regarding the limitation of chip exports to China. This sentiment has been intensified by the recent scandal surrounding Nvidia's leadership. The legislative focus is shifting toward more aggressive measures to protect national security interests.

The think tank Gavekal Dragonomics warned that these developments could lead to a tightening of controls by the end of this year. They highlighted that the scandal, combined with a more AI-skeptical Congress, could make it much harder for Chinese firms to get the computing power they need. This creates a precarious environment for any company building large language models.

Impact on computing and infrastructure

Chinese AI companies currently rely on various methods to obtain the advanced processors required for their operations. Many have been attempting to rent Nvidia-powered computing capacity from offshore cloud providers to bypass direct restrictions. However, new fears suggest that Washington may soon target these remote access points as well.

If the US government successfully restricts the renting of offshore cloud computing power, it would create a major bottleneck for AI development. Analysts suggest that all companies with large language model businesses would be negatively impacted by such a move. The lack of high-end compute power directly limits the ability to train and deploy sophisticated AI systems.

The potential restrictions also extend to the physical components used in data centers. The US government is reportedly working on a measure to bar the import of new Chinese optical transceivers. These components are vital for data transmission, and a ban would deal a significant blow to suppliers who derive most of their revenue from the United States.

Winners and losers in the shift

The market is beginning to differentiate between companies that rely on US imports and those that produce domestic alternatives. While international suppliers face headwinds, homegrown Chinese chipmakers are seeing a surge in investor interest. This is driven by the belief that China must develop its own independent hardware ecosystem to survive potential sanctions.

Cambricon Technology, a Shanghai-listed AI chipmaker, rose 1.2% and was on track for a 22% weekly jump, marking its best performance in six months. Similarly, domestic chipmaker Hygon Information Technology was poised to gain 7% during the same period. These gains reflect a strategic pivot toward supporting local technology infrastructure.

In contrast, companies like Innolight and Eoptolink face a dual threat of reduced demand and potential import bans. Because their business models are so closely tied to the US market, they are particularly vulnerable to the current geopolitical climate. The shift toward self-reliance creates a clear divide between the winners and losers of the current trade tension.

Looking ahead at the AI landscape

The long-term outlook for Chinese AI depends on how quickly the domestic industry can bridge the gap in computing power. If US export controls are tightened as predicted, Chinese firms will have to accelerate their transition to local chips. This transition requires significant investment in manufacturing and research and development.

The situation also underscores the interconnected nature of the global tech supply chain. A scandal involving a single US executive can trigger a sequence of policy changes that affect companies thousands of miles away. The industry is now watching for specific legislative actions from the US Congress regarding chip-making equipment.

Ultimately, the next few months will be critical for determining the pace of AI development in the region. If the US government acts on its current intentions, the landscape for Chinese AI models could become increasingly isolated. For now, the market remains in a state of high alert as it waits for the next move from Washington.